KR101523392B1 - Method of radio data emission, emitter and receiver using the method - Google Patents

Abstract

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a wireless communication domain, and more particularly, to a wireless communication domain, which is useful in a context of multiple input multiple output (MIMO) or multiple input single output (MISO) ) And a decoding technique.

The present invention proposes a space frequency block code (SFBC) based on simultaneous use of two Alamuti codes in the frequency domain, where each Alamuti code is applied to a well-selected frequency pair.

The proposed SFBC maintains a constant envelope characteristic for each antenna while performing the same or very similar performance as pure Aramuti STBC performance.

Wireless communication, chung-yak, alamuti, SFBC, radiator, transmitter

Description

TECHNICAL FIELD The present invention relates to a wireless data radiating method, a radiator and a transmitter using the same,

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to wireless communication areas, and more particularly, to a method and apparatus for performing useful coding and decoding in the context of Multiple Input Multiple Output (MIMO) or Multiple Input Single Output (MISO) and a decoding technique.

Orthogonal Frequency-Division Multiplexing (OFDM) such as Coded Orthogonal Frequency-Division Multiplexing (COFDM) is based on a Frequency Division Multiplexing (FDM) scheme, but is implemented in a digital modulation scheme. The bit-stream to be transmitted is generally divided into dozens to thousands of multiple parallel streams. The available frequency spectrum is separated into a plurality of sub-channels, and each low-rate bit-stream is transmitted on one sub-channel by modulating the sub-carriers using a reference modulation technique such as PSK, QAM,

The sub-carrier frequency is selected so that the modulated data streams are orthogonal to each other to eliminate cross tallk between subchannels. This orthogonality occurs when the sub-carriers are at equal intervals by the symbol rate of the sub-carriers.

The fundamental advantage of OFDM is its ability to cope with multipath and narrowband interference, complex equalization filters in harsh channel environments. Channel equalization is simplified by using a much slower modulated narrow band signal instead of a rapidly modulated wide band signal.

A deformation technique called DFT-spread OFDM has been developed. Each symbol to be transmitted in the system is spread through a frequency transmitted by a DFT (Discrete Fourier Transform), and the resulting signal is transmitted to a conventional OFDM transmission system.

Figure 1 shows a code implementation in the frequency domain for a transmitter. Particularly when a MIMO technique is used and it is desirable for simplicity to implement in the frequency domain to improve the frequency discrimination between emitters emitting in different frequency bands, actual implementations can be done in the frequency domain or time domain have.

The transmitted data is mapped onto a symbol and coded by a coding and modulation module (1.1) which provides a group of symbols (x _n ). The signal is then spread in the frequency domain by a Fast Fourier Transform (FFT) module 1.2. Thereafter, a frequency mapping (1.3) step, which may include zero insertion, frequency shaping, frequency transposition, and possible filtering, such as oversampling in the time domain, is performed.

For the sake of simplicity, the output of the frequency mapping module 1.3 is a vector of N 'size larger than N, even if N and N' assumed below are not equal if there is no loss of generality when including zero insertion Is generated. The signal is the symbol (x _n) and not even very close to the symbol to return to the group _{(x 'n) (giving back} ) transformed back to the time domain by the IFFT (Inverse Fast Fourier Transform) ( 1.4) In the same for transmission do. CP (Cyclic Prefix) insertion (1.5) may be applied before transmission.

Figure 2 illustrates a decoding implementation in the frequency domain for a receiver. The received data is first synchronized in step 2.1. If the coder inserts a CP (Cyclic Prefix), it is removed in step 2.2. The FFT 2.3 is then applied to transform the signal into the frequency domain. Thereafter, an equalization step (2.3) is performed using data according to the channel state obtained by the channel estimation (2.7) step. The data is despread by IFFT 2.5 before the demodulation and channel decoding (2.6) steps.

The system has good properties. In particular, the transmitted signal maintains a constant envelope. When a CP (Cyclic Prefix) is inserted into a signal, it is particularly easy to implement a minimum mean square error (MMSE) linear equalizer in the frequency domain.

It is known that the use of a plurality of antennas at the radiator leading the MISO system or at the radiator and the transmitter leading to the MIMO system is useful for improving the robustness of the transmission.

Such improved robustness can be used to increase range or bandwidth by adjusting the trade-off relationship between range and bandwidth. A diversity scheme can be used with the advantage of a multi-antenna in a radiator.

Alamouti developed the STBC (Space Time Block Code) code because the transmitted information spreads in space by using different time slots (in time) by different antennas in space .

&Quot; Simple Transmit Diversity Technology for Wireless Communication ", IEEE J. select. Areas Commun, vol. 16, pp. 1451-1458, October 1998. " is a reference article on Alamouti code.

In the first implementation of the Alamouti code, two transmit antennas (Tx1, Tx2) and two symbols a and b transmitted in two time slots (T1, T2) are planned, and at time T1 the antenna TX2 transmits symbol b , Antenna TX1 transmits symbol a. When transmitting symbol a * at antenna Tx2 at time T2, antenna TX1 transmits symbol-b *. Here, * denotes a complex conjugate. This is illustrated in FIG.

The above-mentioned Alamouti code has advantages of providing simple coding and decoding. Diversity has been increased to lead to better performance. The maximum A Posteriori (MAP) decoding is very simple, as long as the channel does not vary between T1 and T2, and as long as the channel is characterized by simple multiplication, the matrix inversion, (log enumeration) or SD (Sphere Decoding). Thus, it is well coupled to OFDM or OFDM based modulation techniques.

A second implementation of the Alamouti code called OSFBC (Orthogonal Space Frequency Block Code) is shown in FIG. It is based on the transmission of data for two different frequencies (F1, F2) rather than two different time slots. When antenna Tx2 transmits symbol b with two symbols a and b transmitted on two transmission antennas Tx1 and Tx2 and two frequencies F1 and F2, send.

When antenna TX2 transmits symbol a *, antenna TX1 transmits symbol-b * on frequency F2. Such Alamouti transformations are referred to as general codes in frequency. Traditionally, the two frequencies are close to limit the variation of the channel. By definition, such techniques apply to OFDM or OFDM-based modulation techniques. Includes some frequency domain implementations of single-carrier techniques characterized by OFDM-based modulation, for example (but not necessarily) the addition of a Cyclic Prefix (CP) such as the DFT-spread OFDM described above. do.

Compared to OSTBC, the advantage is that it uses only one modulation slot that can be advantageous from a multiplexing point of view and can lead to better performance in the case of very fast variations of channels such as Doppler. The disadvantage is that the channel can vary between the two frequencies, which can degrade performance or increase the complexity of the receiver. Because of its simple implementation and good performance, Alamuti code is a very attractive technology to be used for MIMO transmission.

Unfortunately, when applied to OFDM or OFDM-based modulation techniques, the Alamouti code does not have an important feature to produce a signal with the envelope characteristic for each antenna, the envelope being the modulus of the complex envelope.

It is an object of the present invention to design a space-frequency code that is easy to code and decode to maintain a constant-envelope characteristic for each transmit antenna with good performance. Preferably, embodiments of the present invention are made in the frequency domain of the receiver stage to simplify decoding. Such an implementation in the frequency domain of the receiver can be used for all transmitters implemented in the frequency domain or time domain of the transmitter.

In order to overcome the above problems, the present invention provides a space frequency block code (SFBC) based on using two different Alamouti codes simultaneously in the frequency domain. The proposed SFBC, each type of Alamouti code applied to a well-selected frequency pair, maintains a constant-envelope characteristic for each antenna, leading to the same or very similar performance as pure Alamouti STBC performance.

The present invention comprises the steps of radiating a signal representing a complex symbol 'X _k ' in a frequency domain on a first antenna for a given time slot on each frequency 'k' (k is 0 to N-1) but k comprises the step of emitting a signal representing the first symbol 'Y _k' on a second antenna during the same time slot on the N-1) at 0,

Here, with respect to 0 and a given constant value selected by the even values between M N-1, the symbol Y _k is derived by the equation below from the symbols X _k for each frequency k,

Y _k = ε (-1), and _{^{K + 1 X * M-1}} -K, where, ε is 1 or -1, X * means the complex conjugate of X and, M-1-K is N modulo ( taken modulo N)

Frequency pairing according to a pairing scheme, applying a modified Alumti code on a symbol transmitted in each frequency pair, and determining if each antenna has at least an even number (strictly, greater than 4) "N & And a method for radiating wireless data by a radiator including at least two transmission antennas for transmitting the wireless data.

In a particular embodiment of the invention, N can be divided by 4, and the value of M selected is N / 2.

In a special embodiment of the invention, the method comprising the N symbols Y _k (N from the symbols X _k according obtaining the N symbols X _k in the frequency domain with respect to the data to be transmitted (N symbols X _k), and the equation calculating the symbols Y _k), and wherein the and from the X _k symbols generating the N signals (N said signals) to be emitted on the first antenna, for from the symbol Y _k is emitted on the second antenna N signals (N said signals).

In a particular embodiment of the invention, the method comprises the steps of obtaining N symbols x _n (N symols X _n ) in the time domain for the data to be transmitted and calculating the formula y _n = 竜 W ^{(M-1) n} x * _{nN / 2} calculating the N symbols y _n (N symols x _n) from the symbol x _N by ^{(where, W = e j2π / N,} x * represents a complex conjugate of x, nN / 2 is given by N in the module.) Generating the N signals (N said signals) emitted on the first antenna from the x _n symbols, and generating the N signals (N said signals) emitted on the second antenna from the y _n symbols .

The invention also relates to a method for decoding at least two transmissions, each comprising a step of applying a spatial frequency code decoding module for each received signal pair in accordance with another allometry encoding code used by a radiator for encoding a signal emitted on a frequency pair To a method for receiving wireless data for signals on at least even frequencies radiated as described above from a transmitter comprising an antenna.

The invention also relates to a method for transmitting at least two different transmit antennas on a first antenna during a given time slot on each frequency 'k' (k is 0 to N-1) Means for emitting a signal indicative of a complex symbol 'X _k ' in the frequency domain, means for generating a complex symbol Y _k (k) on the second antenna during the same time slot on each frequency 'k' And means for emitting a signal indicative of the signal;

For a given integer value M selected as an even number between 0 and N-1 (including 0 and N-1), the symbol _Yk is given by the following equation

Y _k = ε (-1), and _{^{K + 1 X * M-1}} -K, where, ε is 1 or -1, X * means the complex conjugate of X and, M-1-K is N modulo ( taken modulo N);

Lt; RTI ID _{= 0.0} > Xk < / _RTI > for each frequency k.

The present invention also relates to a receiver comprising means for receiving said signal radiated according to Clause 1 from a transmitter comprising at least two transmit antennas to at least even frequencies, means for pairing the received signal on said frequency according to a radiated pairing scheme And means for applying a spatial frequency code decoding module for each received signal pair in accordance with a modified Alamouti encoding code used by the radiator to encode the signal radiated on the frequency pair .

Figure 1 shows an encoding implementation in the frequency domain for a transmitter with one radiating antenna.

Figure 2 illustrates a frequency domain decoding implementation for a receiver with one receive antenna.

Fig. 3 shows an Alamouti code. Fig. 3a shows a time domain, and Figs. 3b, 3c and 3c show Alamouti codes in the frequency domain.

4 shows a first embodiment of two antennas.

5 shows a second embodiment of two antennas.

6 illustrates a structure of an encoder in a frequency domain according to a preferred embodiment of the present invention.

FIG. 7 illustrates a structure of an encoder in a frequency domain according to another preferred embodiment of the present invention.

8 illustrates a structure of an encoder in a frequency domain according to another preferred embodiment of the present invention.

9 illustrates a decoder structure of an apparatus having one receive antenna according to a preferred embodiment of the present invention.

FIG. 10 illustrates a decoder structure of an apparatus having a plurality of reception antennas according to a preferred embodiment of the present invention. Referring to FIG.

Figure 11 illustrates an organigram of a spinning method according to a preferred embodiment of the present invention.

12 shows a block diagram of a receiving method according to a preferred embodiment of the present invention.

BRIEF DESCRIPTION OF THE DRAWINGS The features of the present invention will become more apparent from the following description of exemplary embodiments with reference to the attached drawings.

As already mentioned, the problem addressed here is to implement a spatial frequency block code in a transmission system using at least two transmit antennas.

The object of the present invention is to provide a space diversity technique that provides performance close to Alamouti performance while maintaining a constant-envelope characteristic for signals transmitted on each antenna.

Of course, simplification of execution, especially for decoding, is also an important point of solution.

The first known diversity scheme is called DD (Delay Diversity). This is a very simple multi-antenna transmission scheme. The second antenna transmits a delayed version of the signal transmitted by the first antenna.

One obvious disadvantage is increasing the equivalent channel length recognized by the transmitter. In a system having a CP (Cyclic Prefix), a CCD (Cyclic Delay Diversity) is preferable.

Also, as is known, CCDs can be applied to systems using CP, such as OFDM or DFT-spread OFDM, for example. Each block transmitted by the second antenna is a cyclic rotation of the block transmitted by the first antenna. This allows to increase the diversity while using a very simple demodulator. In the case of DFT-spread OFDM, the two transmitted signals have a positive envelope. However, performance is not as good as that obtained with Alamut code, for example.

3A and 3B show a known modified Alamouti code that is common in the time domain and is common in the frequency domain, and FIGS. 3C and 3D show a slightly different modified Alamouti code. It is a matter of course to those skilled in the art that a modification of Alamouti has the same performance as a conventional one.

Two new variants in the frequency domain can be noted. Here, what is shown in FIG. 3C is referred to as a first modification, and what is shown in FIG. 3D is referred to as a second modification.

Since both variants correspond to the transmitted signal when the MIMO scheme is not used, these two variants ensure that the signal transmitted by Tx1 has a positive envelope.

When applying transmission according to the OFDM-based transmission scheme, a plurality of carriers at different frequencies in an applied frequency band must be transmitted. It is proposed that one of the Alamouti variants is applied to each frequency pair. It can be seen that the inlap will depend on the association technique of different frequencies and the selected Alamouti transform for each pair.

In the first embodiment of the present invention, the association of two variants of Alamouti with DFT-spread OFDM, which is performed in frames of even N frequencies, is as follows.

First, it associates the first frequency of index 0 with the last frequency of index N-1 (Nth frequency). Then, it associates the second frequency (index 1) with the N-1th frequency of index N-2. In this way, the remaining frequencies are associated.

Next, the use of two Alamouti code variants (first variant, second variant) is applied alternately to each associated frequency pair as shown in Fig.

Figure 4 illustrates the technique of the present invention for two antennas (Tx1, Tx2). It is natural that this technique can be described using eight frequencies without considering substantial null sub-carriers that could be added, and can be extended to any even number of frequencies.

Symbols X0 through X7 represent other symbols transmitted at different frequencies via antenna Tx1 at a given time. Using the same numbering for frequencies, it can be seen that frequency F0 using Alamuti code variant 1 is associated with frequency F7 and frequency F2 is associated with frequency F5.

At the same time, frequency F1 using Alamouti code variant 2 is associated with frequency F6, and frequency F3 is associated with frequency F4. Of course, the use of two variants can be used interchangeably.

Arranged as described above is a second antenna for transmitting the symbols ^{_{-X * 7, X * 6,}} -X * 5, X * 4, -X * 3, X * 2, -X * 1, X * 0 to the frequency sequence Which leads to the given transmission.

The symbols are diffused on the frequency while maintaining a constant order compared to the order on the first antenna and it can be proved that the signals transmitted by the second antenna have a positive envelope due to the regular alternation of signs have.

In the frequency domain, this corresponds to the following equation.

Where ε is 1 or -1, and ε is essentially the inverse of the use of two variants of Alamouti.

It should be noted here that for the value of N equal to or less than 4, the formula can lead to a known technique. If the execution of the encoder in the time domain is performed, it does not have _Xk symbols in the frequency domain, but instead has the procedure of _xn symbols in the time domain.

Thus, the given relationship between X _k and Y _k symbols in the frequency domain corresponds to a dual relationship in the time domain leading to the generation of the same signal.

, here,

,

x ^* denotes the complex conjugate of x, and nN / 2 is the modulo N (taken modulo N).

The above equation is valid when the zero insertion optional operation which does not significantly change the forward lock characteristic is ignored. The solution can be applied to all even N number. The associated symbols are mainly separated by frequency, and these frequencies are likely to correspond to different channel responses. This can lead to a slight degradation of performance. And can slightly increase the complexity of the underlying Alamouti decoder.

When the above technique is planned for the uplink of a cellular transmission system, the transmitter is generally implemented in the base station and the corresponding increase in complexity is ignored in this case. If N is a multiple of 4 (N = 4p), frequency separation may be reduced by applying the previous scheme for each half of the frequency multiplex transmission system, as shown in FIG. This leads to the second embodiment of the present invention.

In the same manner as in the first embodiment, in the frequency domain, the second embodiment

에 대응하는 시간 영역에서 듀얼 공식을 이끄는

에 대응된다.here,

Leading to a dual formula in the time domain corresponding to < RTI ID = 0.0 >

.

In general, the representation of the indices such as n-N / 2, N / 2-1-K or M-1-K in the entire specification represents the expressed modulo N.

However, for any M that is an even number between 0 and N-1, the above embodiments (first and second embodiments) can be generalized by the following general technique.

- Frequency representation:

 Here, the Alamouti transformation scheme is performed between carriers k and M-1-k.

- Time representation:

It should be appreciated that both time and frequency implementations lead to the generation of very identical signals to be emitted.

와 링크되는 x_n과y_n 심볼에 근거한 시간 영역 구현을 사용하여 발생된다 하더라도 주파수 영역에서 해석된다면, 수학식

, 와 링크되고, X_k와 Y_k 심볼의 표현으로서 인식될 수 있다.In the time domain,

If it is interpreted in the frequency domain, even if it is generated using a time domain implementation based on the x _n and y _n symbols linked with equation

, And can be recognized as a representation of X _k and Y _k symbols.

It should also be noted that the above equations do not take into account the frequency shift and interpolation that can be performed on frequency or time.

The first embodiment corresponds to M = 0, and the second embodiment corresponds to M = N / 2 which implies that N / 2 is an even number. Therefore, N = 4p. All embodiments solve the problem for achieving performance close to Alamouti performance, while maintaining the normal-grained nature.

Implementation according to a preferred embodiment of the present invention will now be described.

A first variant on the frequency implementation of the encoder is provided in Fig. 6, and a second variant on the frequency implementation is provided in Fig. The data to be transmitted is coded by the coding and modulation module 6.1, 7.1, which provides a group of symbols x _n , and mapped onto a symbol.

The signals are then spread in the frequency domain by Fast Fourier Transform (FFT) modules 6.2 and 7.2. And perform frequency mapping modules 6.3 and 7.3, which may include zero insertion, frequency shaping, and so on.

The signal is converted back to the time domain by an Inverse Fast Fourier Transform (IFFT) module (6.4, 7.4) which returns the symbol group x _n 'very close to the original symbol x _n , even if it is not the same for transmission. An optional CP insertion (6.5, 7.5) can be applied before transmission. The second antenna is the first branch and is fed with data calculated by a spatial frequency block code calculator (6.6, 7.6) leading to a new branch with optional CP insertion (6.8, 7.8) and IFFT (6.7, 7.7).

The difference between the two runs is only a disposition of the spatial frequency coding function (6.6, 7.6) for the frequency mapping functions (6.3, 7.3). It should be noted that the implementations are very generic and can be applied to 2X2 spatial frequency coding.

The core of the present invention is a 2X2 spatial frequency block code used in function (6.6, 7.6) substantially according to the given equation above.

An implementation in the time domain of the encoder is shown in Fig. The data to be transmitted is coded and mapped onto a symbol by a coding modulation module 8.1 which provides a group of symbols X _n . Thereafter, a CP insertion (8.5) may optionally be applied to the signal (data). A frequency shaping module (8.9), which may then include oversampling, filtering, and frequency potential prior to transmission, operates.

Module (8.6) applies the time implementation of the code. The module is based on the same equation in the time domain given above. The second branch having the selective CP insertion (8.8) and the oversampling and frequency shaping module (8.10) has the same operation as the first branch.

An implementation of the decoder in the frequency domain for one receive antenna is shown in Fig. 9, and an implementation for multiple receive antennas Nr is shown in Fig.

In Fig. 9, the received data is first synchronized in module 9.1. If the coder has inserted the CP, it is removed from the module (9.2). The FFT 9.3 is then applied to transform the signal in the frequency domain. Then, the equalization module 9.4 performs an operation using the data according to the channel state obtained by the channel estimation module 9.7.

The channel estimation is done for each radiating antenna and each frequency used. The equalization module applies the data to spatial frequency block decoding in accordance with the code used by the encoder. The data is then despread by the IFFT 9.5 before the demodulation and channel decoding module 9.6.

FIG. 10 shows a decoder structure when a plurality of receive antennas are used. A plurality of signals 10.8 are received from the receive antenna. The synchronization module 10.1 synchronizes all these signals. If a CP removal (10.2) is optionally made and used, an FFT (10.3) is also performed in parallel with all synchronized signals before applying each signal.

The Nr module of the channel estimate 10.7 includes an NrX2X2 basic space frequency block decoder that processes N / 2 pairs of subcarriers (one complex module is possible) and Nr Signal. Before general channel decoding (10.6), the IFFT module (10.5) handles the resulting signal.

In order to obtain the advantage from the features of the present invention, decoding is preferably performed in the frequency dimension.

The present invention relates to a spatial frequency block code decoding module which is partially dependent on the spatial frequency block code decoding module itself but which also processes only one carrier pair at a time and the other carrier pairs are decoded one from each other independently Lt; RTI ID = 0.0 > decode < / RTI >

This last property is not achieved by most techniques that would have been defined in the time domain within a module. On the other hand, the classic Alamouti technique performed on the adjacent carrier will also have this property, but it does not have the normal-fork properties.

With respect to the complexity of the SFBC coding shown in FIG. 10 in the case of a multi-antenna, a Minimum Mean Square Error (MMSE) decoder will have a matched filter. This means multiplication by a complex matrix of 2 x 2Nr or an actual matrix of 4 x 4Nr, followed by a solution of a 2x2 complex linear system or a solution of a 4x4 real linear system. More research and simulation will give us confidence that the invention is necessary. In any case, the response complexity is moderate.

Now let's take a closer look at the decoding of MIMO codes for non-stationary channels.

First, a general method for decoding LD (Linear Dispersion) in the real domain is introduced, and then an Alamuti code for a non-static channel and a method applied to its transformation are introduced.

The following parameters are defined.

K is the number of information symbols per code word,

• N is the number of sub-

● L is time or frequency domain. Means the number of time slots or subcarriers associated with spatial or spatial frequency coding. In the case of Alamuti, L is 2.

Nr is the number of receive antennas,

• Nt is the number of transmit antennas.

We define the following vectors and matrices.

● x is a K × 1 complex vector representing information data.

● Xr is a 2K × 1 real vector representing information data in real representation.

• H is a complex matrix of L.Nr × L.Nt size representing the channel response in the frequency domain.

Hi is a complex matrix of Nr x Nt size representing the channel response for time slot i or sub-carrier i.

● Hr is a complex matrix of size 2L.Nr × 2L.Nt representing the channel in the real representation.

• s is a complex matrix of Nt × L size that represents the coded data in the representation of the matrix.

● sv is the vector expression: sv = vect (s), and Nt.L × 1 complex vector representing the coded data.

• A and B are complex matrices of L.Nt × K size representing the encoding matrix.

● Cr is an integer matrix of size 2L.Nt × 2K that represents an equal-thread encoding matrix.

● y is a complex vector of L.Nr × 1 size representing the received data in the vector representation.

● yr is a 2L.Nr × 1 real vector representing the received data in vector and real representation.

● v is a complex vector of L.Nr × 1 size representing the noise vector.

● vr is a 2L.Nr × 1 real vector representing the noise vector in the real representation.

In the conventional description of the MIMO technique, it is assumed that the influence of the channel can be modeled by a linear transformation. In particular, in a broadband system, it is assumed that this technique is performed in the frequency domain. In an OFDM or DFT-spread OFDA system or some special multi-carrier system, this implies that the MIMO scheme can be applied to a small number of subcarriers in the case of STBC or in the case of SFBC.

At this point, STBC and SFBC techniques should be considered separately.

In the STBC scheme, for each subcarrier k of the transmitted server-carrier N, a space-time (ST) encoding is applied and spreads through a generally L time slot in general. Here, a time slot corresponds to an OFDM symbol.

For example, assuming that there are two transmit antennas Tx0, Tx1, and Nt = 2, and L = 2 time slots, symbol a ^k _{i, j} is transmitted on antenna i for time slot j for subcarrier k will be.

In the matrix concept, this corresponds to the transmission of a matrix as follows;

As long as ST encoding and decoding are involved, the corresponding N flows are processed in parallel: therefore, to simplify (WLOG) the loss of generality, the superscript k will be omitted from the following.

In the SFBC scheme, only one time slot (e.g., one OFDM symbol) is involved in a particular spatial frequency (SF) encoding or decoding. However, the N / L SF encoding / decoding will be processed in parallel and individually. Is spread on L different subcarriers, each SF encoding.

The index of the SF encoding is referred to as k. (Where K is between 0 and N / L-1).

The encoding includes L subcarriers k ₀ , k ₁ , ... . , kL _-1 . Lt; / RTI >

Then, after SF encoding, the symbols a ^k _i, ^k _j are transmitted on subcarrier ^k _j by transmit antenna i.

Assuming, for example, that there are two transmit antennas Tx0, Tx1 (Nt = 2) and a subcarrier with L = 2 is used for each SF encoding, this corresponds to transmitting the following matrix;

As long as SF encoding and decoding are involved, the corresponding N / L flows are processed in parallel. Thus, in order to simplify without loss of generality, the superscript k will be omitted from the following. It is considered that the matrix S is transmitted.

The coded data of a space-time or space-frequency (ST / SF) code can be expressed in matrix or vector form. As used above, the more natural way to represent the encoding is in matrix form. Indeed, this matrix type generally leads to an easier representation of the encoding process.

However, the vector form will be used because the vector form simplifies the representation of the channel and the representation of the decoding. In the following, a vector is obtained from a matrix in which other columns are superimposed.

For example, if s and sv are matrix and vector representations of coded data, then

,

,

This can be expressed by the SFBC case by the following.

With a very common sense, the ST / SF encoding can be represented by the following complex representation of the LD code:

sv = Ax + Bx *

Here, x is information under the vector form before ST / SF encoding.

An example of the A and B matrices is provided as follows when expressing the decoding for Alamouti. If B = 0, the ST / SF encoding is linear, which is not the case for the Alamouti technique.

Let's take a look at the channel representation. It is assumed that, in the frequency domain, the channel can be modeled by a simple multiplicative coefficient at a given instant and at a given frequency (e.g., a given sub-carrier).

In the case of SISO, which is the sub-carrier i, the received samples are as follows.

y _i = H _i a _i + noise

Where H _i is the complex coefficient in SISO and a _i is the transmitted value.

The formula for the case of MIMO can be derived directly from the model. For example, in the case of SFBC, samples received at sub-carrier i and receive antenna j

The following equation is obtained.

Here, H _i (j, i) is a complex coefficient corresponding to the frequency channel response between antennas 1 and j at frequency _i , and a _{k, i} is the data transmitted at sub-carrier i by transmit antenna k.

Thus, the channel can be expressed in the following matrix form.

Here, matrix H _i is the frequency response of the channel at time i (STBC case) or frequency i (sub-carrier k _i , SFBC case). And, when the entry (j, l) of the matrix H _i corresponds to the channel coefficient between the transmit antenna l and the receive antenna j, if the channel is stationary, all H _i matrices are the same.

This leads to the fact that the received complex vector is equal to the following equation.

y = Hsv + v = H (Ax + Bx *) + v,

Where v is an additional white Gaussian noise.

The above equation has the same real expression as the following.

yr = HrCrxr + vr

The above equation has a great advantage to be in a linear form.

The vector xr (resp.yr, vr) can be obtained from x (resp. Y. V) by overlapping the real and imaginary regions of the original complex vector.

For example,

Here, superscripts R and I are indications for real and imaginary regions, respectively.

The matrices Cr and Hr can be easily obtained as follows.

And

Quite often, a simple representation of the Minimum Mean Square Error (MMSE) decoding may be found in real or complex form, especially for the Alamouti code.

The complex representation seems to be actually simpler. However, it does not mean that it corresponds to less operation. In addition, the general form of the encoder is not linear in complex form. And a simple complex expression implies, for example, an ad hoc transformation that rely on the code when it exists.

To keep the generalization, we will only express the description of the MMSE decoder only in the real domain, keeping in mind that there is also a simple representation in the complex domain for Alamouti.

The above description of Alamouti codes in the complex domain is provided below.

From the above, the MMSE formula is very simple. The receiver room vector can be reformulated as follows.

Yr = F xr + vr, where F = Hr Cr

This leads directly to the MMSE estimator of xr as follows.

Where s ² is the variation of the complex noise v, I _2K is the unitary matrix of size 2K, and F ^T is the transpose of F.

Whatever the number of antennas and the time slot (or the number of sub-carriers), it is significant that only a 2K-sized real matrix must be inversely transformed. If a complex representation is possible, the 2K complex matrix is inversely transformed.

In the case of Alamouti, it has Nt = 2, K = 2, L = 2, and a variable Nr.

A conventional time or frequency version of Alamouti can be obtained as follows.

The first variant of Alamouti can be obtained as follows.

A second modification of Alamouti can be obtained as follows.

We have described the general formula of an MMSE decoder for all LD codes in the real world.

They can be applied to Alamouti codes and Alamuti code variants with the A and B matrices. Now, the same or simpler representation of the MMSE decoder for the Alamouti code and its variants in the complex domain will now be described.

For each version of Alamouti code, the MMSE decoder may be described in a complex form.

For a generic Alamouti using the above concept, the transmitted vectors are as follows.

The received vector y is as follows.

Where y ₀ and y ₁ are vectors of size Nr.

Y 'is defined by conjugating the data corresponding to the second sub-carrier for the second time slot or frequency code.

The following expression can be directly obtained.

Since this is a linear expression, the MMSE estimate of x is simply provided by the following equation.

Here, D ^H represents Hermitian of D.

For a first variant of Alamouti at the frequency, the transmitted vector is as follows.

Here, x 'is defined as follows.

Now, the transmitted vector is as follows.

After conjugating the second part (second time slot or second subcarrier) as in the previous section, the received vector is expressed as:

Since it is also a complex linear expression, the x 'MMSE estimation is simply provided as follows.

And the MMSE estimation of x is as follows.

For a second variant of Alamouti at the frequency, the transmitted vector is:

X " is defined as follows.

The transmitted vector is now as follows.

And after conjugating the second part as in the previous section, the received vector is now expressed as:

Since it is also a complex linear expression, the MMSE estimation of x " is simply provided as follows.

And the MMSE estimation of x is as follows.

11 shows an organigram of a radiation method in the frequency domain according to a preferred embodiment of the present invention. Step 11.1 is the step of obtaining N symbols x _k in the frequency domain for the data to be transmitted.

Step 11.2 is a step of calculating N symbols Y _k from symbol X _k according to the formula Y _k =? (-1) ^{K + 1} X * _M-1-K . Step 11.4 is the step of generating a signal to be radiated on the first antenna from the X _k symbol, N ', which is greater than N or N in the case of zero insertion, as already mentioned.

Step 11.3 is a step of generating a signal N or N 'to be radiated on the second antenna from the Y _k symbols. In step 11.6, a signal representing the x _n symbol is radiated on the first antenna. On the other hand, a signal representing the y _n symbol is emitted on the second antenna in step 11.5.

12 is a block diagram of a receiving method in a case where one receiving antenna is provided according to a preferred embodiment of the present invention. Step 12.1 is the step of obtaining N received symbols in the frequency domain for the transmitted data. Step 12.2 is a step of applying to the Alamouti decoding module for each received pair signal according to a variation of the Alamouti encoding code used by the emitter to encode the signal emitted on the frequency pair. In step 12.3, an estimate of x _n symbols in the time domain is obtained from the estimated symbols in the frequency domain.

The MIMO technique described with two antennas can be extended to any even numbered antenna. The expansion is done by grouping the antennas in pairs. And applying one of the coding schemes as defined in the present invention to each antenna pair.

The present invention can be applied to any transmission system that uses multiple transmitters. Although it can be used in line transmission where crossover interference occurs, the present invention is likely to be closely related to the wireless system.

Furthermore, the present invention is described in a DFT-spread OFDM context. However, even if the actual advantage lies solely in the modulation with a positive envelope, the proposed invention can be used in any modulation technique.

Regarding CP (Cyclic Prefix), the present invention simplifies receiver implementation in the frequency domain. However, even though it may be more complicated, other frequency domain implementations of the transmitter may be possible without the CP.

An example of such an implementation is an overlap method (e.g., OLA, OverLap Add, or OLS, OverLap Save).

In the OLS method, N samples will be handled in the frequency domain and will be transformed into the time domain. And only a few of them will be kept in the time domain. The corresponding handling window is overlapping to ensure that all received samples are handled.

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to wireless communication areas, and more particularly, to a method and apparatus for performing useful coding and decoding in the context of Multiple Input Multiple Output (MIMO) or Multiple Input Single Output (MISO) and is applicable to all communication systems for data transmission.

Claims (8)

A wireless data radiated method by a radiator comprising at least two transmit antennas, each antenna transmitting on at least an even number 'N' (preferably greater than 4) different frequencies, Transmitting a signal representing a complex symbol 'X _k ' in the frequency domain on the first antenna during a given time slot on each frequency 'k' (where k is 0 to N-1); Transmitting a signal indicative of a symbol Y _k on the second antenna during each said time slot on each frequency 'k' (where k is 0 to N-1) With respect to the zero and an integer value M chosen as an even number of the given N-1 between (with 0 and N-1), is derived from the symbols X _k for each frequency k by the symbol Y _k is the formula to, Y _k =? (-1) ^{K + 1} X ^* _M-1-K , Here, ε is 1 or -1, X ^* means the complex conjugate of X and, M-K-1 is a module N (taken modulo N), and the index k and the index M-1-k according to the pairing method Is paired, and a transform Alamouti code is applied to each frequency pair for the symbol to be transmitted. The method according to claim 1, N is divided by 4, and the value of the selected M is equal to N / 2. The method according to claim 1, Wherein the value of the selected M is equal to zero. 4. The method according to any one of claims 1 to 3, Obtaining the N symbols X _k (N symbols X _k) in the frequency domain for the data to be transmitted, and Calculating the N symbols Y _k (N symbols Y _k) from the symbols X _k according to said formula, and Generates N said signals to be transmitted from the X _k symbol to a first antenna, And generating the N signals (N said signals) to be transmitted from the Y _k symbol to a second antenna. 4. The method according to any one of claims 1 to 3, Acquires N symbols x _n in the time domain with respect to the data to be transmitted, Calculates N symbols y _n from the symbol x _n according to the following equation, ^{_{y n = εW (M-1}} ) n x * nN / 2 ( ^{where, W = e j2π / N,} x * represents a complex conjugate of x, nN / 2 is a modulo N) Generating the N signals to be emitted on the first antenna from the x _n symbols and generating the N signals to be emitted on the second antenna from the y _n symbols, Radiation method. A method for wireless data reception of a signal on at least one even frequency, said signal to be transmitted in accordance with claim 1, from a transmitter comprising at least two transmit antennas, Pairing the received signal on the frequency according to a radiation pairing scheme; Applying a spatial frequency code decoding module to each received signal pair in accordance with a modified Alamouti encoding code used by the radiator to encode the signal radiated on the frequency pair. . At least two transmit antennas; Means for transmitting on each antenna at least at even N different frequencies; Means for emitting a signal indicative of a complex symbol X _k in the frequency domain on a first antenna during a given time slot on each frequency 'k' (k is 0 to N-1); Means for emitting a signal indicative of a symbol Y _k in the frequency domain on the second antenna during each said time slot on each frequency 'k' (k is 0 to N-1); With respect to the zero and an integer value M chosen as an even number of the given N-1 between (with 0 and N-1), the symbol Y _k is derived from the symbols X _k, for each frequency k by the formula below, Y _k =? (-1) ^{K + 1} X ^* _M-1-K , Here, ε is 1 or -1, X ^* means the complex conjugate of X and, M-K-1 is a module N (taken modulo N), and the index k and the index M-1-k according to the pairing method Is paired, and a transform Alamouti code is applied to each frequency pair for the symbol to be transmitted. Means for receiving said signal radiated according to claim 1 on at least even frequencies from a transmitter comprising at least two transmit antennas; Means for pairing received signals on the frequency according to a radiation pairing scheme; Means for applying a spatial frequency code decoding module for each received signal pair in accordance with another Alamouti encoding code used by the emitter to encode the signal radiated on the frequency pair.

Images